Keywords: reinforcement learning, continual learning, universal value functions, off-policy learning, multi-task
TL;DR: Agents learning jointly and off-policy about many tasks make progress on challenging continual learning domains.
Abstract: Some real-world domains are best characterized as a single task, but for others this perspective is limiting. Instead, some tasks continually grow in complexity, in tandem with the agent's competence. In continual learning there are no explicit task boundaries or curricula. As learning agents have become more powerful, continual learning remains one of the frontiers that has resisted quick progress. To test continual learning capabilities we consider a challenging 3D domain with an implicit sequence of tasks and sparse rewards. We propose a novel agent architecture called Unicorn, which demonstrates strong continual learning and outperforms several baseline agents on the proposed domain. The agent achieves this by jointly representing and efficiently learning multiple policies for multiple goals, using a parallel off-policy learning setup.